38 INTRODUCTION. 



drawing any inference from facts, which, though true, 

 have been noticed but in a limited number of in- 

 stances. We must enter upon the study of Nature 

 prepared to believe everything which is not actually 

 impossible, and which has not certainly been dis- 

 proved ; ready to confirm the observations, however 

 strange, of those who have gone before us, if occasion 

 offer ; ready, also, to give up what we had considered 

 our best established principles, if called upon to do 

 so by a more extended research. In a word, we 

 must equally repress a harsh judgment upon the 

 statements of others, and a too hasty desire to 

 generalize on our parts. 



(24.) Besides observing the same fact in a suffi- 

 cient number of instances, we should, as far as 

 possible, observe it in instances, in which the attend- 

 ing circumstances are different. This is especially 

 of consequence in all that relates to the habits of 

 animals, and the dates of periodic phenomena ; for 

 these are often so modified by some peculiarities of 

 place or season, as to offer quite a different result 

 from what would ordinarily occur. To mention 

 many examples of either of these two classes of 

 exceptions would be unnecessary. As regards the 

 manner in which instinct often conforms to cir- 

 cumstances, we may just allude to the case of birds 

 choosing such materials for their nest as are most 

 readily obtained, or adapting the situation of it to 

 the nature of the particular locality.* The altered 



* See White's Selborne, Lett. 56, to Daines Barrmgton ; also 

 Lett. 21 to Pennant, on the subject of daws breeding in unlikely 

 spots. See also Bolton's Harmonia Ruralis, vol i. p. 492, 



